The OBSERVER is a video surveillance system that detects and predicts abnormal behaviors aiming at the intelligent surveillance concept. The system acquires color images from a stationary video camera and applies state of the art algorithms to segment, track and classify moving objects. In this paper we present the behavior analysis module of the system. A novel method, called dynamic oriented graph (DOG) is used to detect and predict abnormal behaviors, using real-time unsupervised learning. The DOG method characterizes observed actions by means of a structure of unidirectional connected nodes, each one defining a region in the hyperspace of attributes measured from the observed moving objects and having assigned a probability to generate an abnormal behavior. An experimental evaluation with synthetic data was held, where the DOG method outperforms the previously used N-ary trees classifier

Consumer-grade virtual reality systems have become increasingly accessible over the last years, making these great options for psychological and physiological medical use. This paper studies the precision of one available system, the Oculus Quest 2. We investigated studies that approached testing of these types of systems using manual systems and automated systems using robot arms and decided to use the latter method for our evaluation. A setup was created where the robotic arm would perform diverse exercises, with the Quest controller attached to it while the headset was either stationary or being worn by a participant. The results show that these systems are precise enough to measure movements that would not be noticed by therapists during traditional rehabilitation and are therefore adequate for medical use.

This paper describes a new approach to perform segmentation of moving objects in real-time from images acquired by a fixed color video camera and is the first tool of a major project that aspires to recognize abnormal human behavior in public areas. The moving objects detection is based on background subtraction and it is unaffected by changes in illumination, i.e., cast shadows and highlights. Furthermore it does not require a special attention during the initialization process, due to its ability to detect and rectify ghosts. The results show that with image resolutions of 380/spl times/280 at 24 bits per pixel, the time spent in the segmentation process is around 80 ms, in a 32 bits 3 GHz processor based computer.

In this paper, we explore data collected in a pilot project that used a digital camera and facial recognition to detect foot traffic to a sports store. Using a time series approach, we model daily incoming store traffic under three classes (all faces, female, male) and compare six forecasting approaches, including Holt-Winters (HW), a Support Vector Machine (SVM) and a HW-SVM hybrid that includes other data features (e.g., weather conditions). Several experiments were held, under a robust rolling windows scheme that considers up to one week ahead predictions and two metrics (predictive error and estimated store benefit). Overall, competitive results were achieved by the SVM (all faces), HW (female) and HW-SVM (male) methods, which can potentially lead to valuable gains (e.g., enhancing store marketing or human resource management).